Over the last year, a method has been developed for continuous-flow scanning of selected high performance liquid chromatography (HPLC) peak components by microprocessor control and applied to the analysis of extracts from human lymphocytes. Continuous-flow wavelength scanning of compounds separated by HPLC is achieved through the use of fixed and variable wavelength micro-ultraviolet detectors connected in series, but separated by a low pressure three-way valve. Activation of the valve allows entrapment of selected peaks in the variable wavelength detector without interfering with the response of the fixed wavelength detector that is utilized for peak quantitation. A microprocessor program is employed to maintain control and accuracy during the scanning sequence. Good correlation was found between ultraviolet spectra of standards obtained on a conventional spectrometer and those on separated peaks. This system allows the identification and quantitation of picomole (pmole) amounts of peaks separated during one analysis of a biological sample. Studies on the development of a HPLC assay for the measurement of dehydroascorbate levels in lymphocytes were also performed. Dehydroascorbate (DHA) can be separated from ascorbic acid via HPLC. However, for lymphocytes, the presence of other nucleosides and bases that elute with overlapping retention times, as well as for the far UV and low extinction coefficient of DHA, necessitated the development of a more sensitive means for DHA determination. This has been achieved through the reduction of DHA to ascorbic acid, employing dithiothreitol. A method recently has been described that suggests conversion of DHA to ascorbic acid by dithiothreitol addition to a sample that already contains ascorbic acid and comparison of the differences before and after the dithiothreitol addition. We have isolated the HPLC fractions containing DHA from lymphocyte extracts, carried out the dithiothreitol reduction with the addition of dithiothreitol to a concentration of 20 mM in a reduced volume (250 microliters) at an optimal pH 6.5 and reanalyzed the sample by HPLC. Since the conversion of DHA to ascorbic acid has been found to be essentially 90%, detection of DHA in a sample can be determined at the same lower limit of sensitivity (approximately 30 pmoles) as obtained with ascorbic acid. However, since DHA using the Roe assay represents about 15 to 20% of the total ascorbate DHA levels in normal lymphocytes, an extract of 1.5 x 10[unreadable]7[unreadable] cells is required to allow reliable measurement of DHA via reduction by dithiothreitol. (B)